The life expectancy of people with serious mental illness (SMI)-schizophrenia spectrum disorders, bipolar disorders-is about 25 years shorter than that those without SMI. This early mortality is due largely to preventable chronic conditions, including those caused or worsened by the metabolic side effects of medications prescribed for SMI. In addition to being at increased risk for life-threatening chronic conditions, patients with SMI may be less likely than those without SMI to receive the guideline-concordant preventive care for these disorders. Little is known, however, about overall patterns of preventive care service use among patients with SMI, or about modifiable factors at the patient, clinician, organizational, or regulatory level that affect uptake of preventive services. New initiatives to address health disparities experienced by individuals with SMI have been proposed, including creating medical care homes within psychiatric clinics and integrating medical and psychiatric care; yet it is unclear if such strategies represent the best approaches, given the dearth of research assessing the underlying factors that produce these disparities. We propose a 5-year, mixed methods study to examine the role of modifiable regulatory-, organizational-, provider-, provider-patient-, and patient-level factors contributing to disparities in preventive service use among patients with serious mental illness. We use electronic medical record data from a large non-profit integrated health plan and an equally large network of safety-net clinics to examine the preventive service use of approximately 65,000 individuals with SMI (age 18 and over), and compare them to similarly aged individuals without SMI (n > 700,000) to identify patient, provider, organizational, and regulatory predictors of preventive service use. We will also conduct a small number of formative, semi-structured interviews with patients and providers to develop a clinician survey and patient interview materials (structured and semi- structured). Information from clinician surveys will be used to predict preventive service use at the panel level. Semi-structured interviews with patients will be used to identify factors affecting willingness to seek preventive services and barriers to, and facilitators of, such service seeking. This mixed methods study will employ triangulated, complementary, methods to gain a nuanced understanding of factors responsible for disparate and inadequate preventive service use among SMI patients, and to identify possible methods for ameliorating barriers to service use. This information will inform ongoing efforts to improve delivery of medical care to individuals with SMI as well as initiatives targeting excess mortality in this population.